Weather Stations
The middleware provides mechanisms for logging performance from two sources. First, the WeatherStations provide the ability to periodically run network tests to measure key metrics. Second, applications log their actual performance experience as they operate. Both sources are placed into a common SQL database. These sources of data are then correlated to provide information on an applications performance experience when certain network conditions prevail. Applications can use this information to both adapt their behavior and provide users with an expectation of performance as compared to previous experience.
Network performance is monitored by WeatherStations that make use of the InformationChannels framework. To monitor end-to-end performance across the national testbeds used in the HAVnet project we have operated weather stations at Stanford, CENIC at Sunnyvale, WISCNet at Milwaukee, UW La Crosse, Athey Lab in Ann Arbor, Michigan, and the Univ. of Michigan Medical School.
Each weather station announces a channel indicating its ability to conduct various network measurements. By querying the registration server each weather stations discovers the existence of peer stations and schedules tests. The tests consist of
- Ping loss and roundtrip time
- TCP throughput
- UDP maximum send rate with less then 1% loss
- Multicast group setup time and loss rate
Weather stations are periodically polled for the results, which are placed into the performance database as well as used to update performance graphs.
